diff --git a/README.md b/README.md
index d6275c0..ac5f7e6 100644
--- a/README.md
+++ b/README.md
@@ -387,17 +387,15 @@ enrichment:
   enabled: true
   incremental: true
   refresh_mode: "manual"
-  providers: [tmdb, douban]
+  providers: [tmdb]
   cache_db: "~/.vlm/enrichment_cache.db"
   max_concurrency: 6
   min_match_score: 0.75
-  douban_endpoint: null
   translation:
     mode: "bidirectional"
     fallback_machine: true
   api_keys:
     tmdb: null
-    douban: null
     openai: null
   reputation:
     min_votes: 50
diff --git a/REVIEW_REPORT.md b/REVIEW_REPORT.md
index 415e4d5..5e55c7d 100644
--- a/REVIEW_REPORT.md
+++ b/REVIEW_REPORT.md
@@ -218,6 +218,56 @@
   - Add developer note for test import convention: always import from package root (`vlm`), not `src.vlm`.
 
 ### Execution checklist (single pass)
+
+## Enrichment 增量评审报告（2026-02-10）
+
+### 评审范围
+- 文件：`src/vlm/enrichment.py`
+- 关注点：刷新一致性、配置可观测性、命名输出质量
+
+### 总体评分（修复后）
+- 当前得分：**92 / 100**
+- 评分依据：
+  - 正确性（40 分）：37/40（refresh 无命中不再残留旧字段）
+  - 健壮性（30 分）：27/30（未知 provider 已 fail-fast）
+  - 输出质量（20 分）：18/20（display title 去重）
+  - 可维护性（10 分）：10/10（TMDB-only，分支复杂度降低）
+
+### 问题明细与修复状态
+
+#### 1) P1 - 刷新失败时未清理陈旧富化字段（**fixed**）
+- 修复位置：`src/vlm/enrichment.py:378`
+- 修复内容：
+  - `_apply_payload` 从“仅写入非 None”改为“payload 含 key 即覆盖”，支持将字段显式清空为 `None`。
+- 验证：
+  - `tests/test_enrichment.py:187` 覆盖 refresh 后无匹配场景，断言旧 `canonical_id/title_zh/title_en/reputation_*` 被清空，`enrichment_confidence == 0.0`。
+
+#### 2) P2 - 未识别 provider 被静默忽略（**fixed**）
+- 修复位置：
+  - `src/vlm/enrichment.py:157`
+  - `src/vlm/config.py:271`
+- 修复内容：
+  - `_build_providers` 对未知 provider 直接抛 `ValueError`（fail-fast）。
+  - 配置校验新增白名单，仅允许 `tmdb`。
+  - 同步重构为 TMDB-only，移除 enrich 主流程中的 Douban 分支。
+- 验证：
+  - `tests/test_enrichment.py:176` 覆盖未知 provider 报错。
+  - `tests/test_config.py:447` 覆盖配置校验拒绝不支持 provider。
+
+#### 3) P2 - display_title fallback 可能重复标题（**fixed**）
+- 修复位置：`src/vlm/enrichment.py:423`
+- 修复内容：
+  - `_build_display_title` 增加同值去重逻辑；`title_zh/title_en` 回退同值时只保留一个。
+- 验证：
+  - `tests/test_enrichment.py:260` 断言双空回退时输出单标题而非重复标题。
+
+### 结果
+1. 三项问题均已修复并有测试覆盖。
+2. enrich 已切换为 TMDB-only，配置和文档已同步。
+3. 全量测试通过：`uv run --with pytest --with hypothesis pytest -q` -> `411 passed`。
+
+### 修复后目标分
+- 达成：**92 / 100**。
 1. Apply Phase 1, run focused + full tests.
 2. Apply Phase 2, add timezone tests, run focused + full tests.
 3. Apply Phase 3, run scanner tests + full tests.
diff --git a/src/vlm/config.py b/src/vlm/config.py
index 61d426c..72fa6f1 100644
--- a/src/vlm/config.py
+++ b/src/vlm/config.py
@@ -30,15 +30,13 @@ class Config:
     enrichment_enabled: bool = True
     enrichment_incremental: bool = True
     enrichment_refresh_mode: str = "manual"
-    enrichment_providers: list[str] = field(default_factory=lambda: ["tmdb", "douban"])
+    enrichment_providers: list[str] = field(default_factory=lambda: ["tmdb"])
     enrichment_cache_db: Path = field(default_factory=lambda: Path.home() / ".vlm" / "enrichment_cache.db")
     enrichment_max_concurrency: int = 6
     enrichment_min_match_score: float = 0.75
     translation_mode: str = "bidirectional"
     translation_fallback_machine: bool = True
     tmdb_api_key: Optional[str] = None
-    douban_api_key: Optional[str] = None
-    douban_api_endpoint: Optional[str] = None
     openai_api_key: Optional[str] = None
     reputation_min_votes: int = 50
     reputation_low_score_threshold: float = 6.0
@@ -103,7 +101,7 @@ def load_config(path: Path) -> Config:
         enrichment_enabled=enrichment.get("enabled", True),
         enrichment_incremental=enrichment.get("incremental", True),
         enrichment_refresh_mode=enrichment.get("refresh_mode", "manual"),
-        enrichment_providers=enrichment.get("providers", ["tmdb", "douban"]),
+        enrichment_providers=enrichment.get("providers", ["tmdb"]),
         enrichment_cache_db=Path(
             enrichment.get("cache_db", str(Path.home() / ".vlm" / "enrichment_cache.db"))
         ).expanduser(),
@@ -112,8 +110,6 @@ def load_config(path: Path) -> Config:
         translation_mode=translation.get("mode", "bidirectional"),
         translation_fallback_machine=translation.get("fallback_machine", True),
         tmdb_api_key=api_keys.get("tmdb"),
-        douban_api_key=api_keys.get("douban"),
-        douban_api_endpoint=enrichment.get("douban_endpoint"),
         openai_api_key=api_keys.get("openai"),
         reputation_min_votes=reputation.get("min_votes", 50),
         reputation_low_score_threshold=reputation.get("low_score_threshold", 6.0),
@@ -155,10 +151,8 @@ def create_default_config(path: Path) -> Config:
             },
             "api_keys": {
                 "tmdb": default_config.tmdb_api_key,
-                "douban": default_config.douban_api_key,
                 "openai": default_config.openai_api_key,
             },
-            "douban_endpoint": default_config.douban_api_endpoint,
             "reputation": {
                 "min_votes": default_config.reputation_min_votes,
                 "low_score_threshold": default_config.reputation_low_score_threshold,
@@ -276,6 +270,13 @@ def validate_config(config: Config) -> list[str]:
         errors.append("enrichment_cache_db must be a Path object")
     if not isinstance(config.enrichment_providers, list) or not config.enrichment_providers:
         errors.append("enrichment_providers must be a non-empty list")
+    else:
+        allowed_providers = {"tmdb"}
+        invalid = [provider for provider in config.enrichment_providers if provider.lower() not in allowed_providers]
+        if invalid:
+            errors.append(
+                f"enrichment_providers contains unsupported providers: {invalid}; supported providers: ['tmdb']"
+            )
     if config.enrichment_max_concurrency < 1:
         errors.append("enrichment_max_concurrency must be >= 1")
     if not (0.0 <= config.enrichment_min_match_score <= 1.0):
diff --git a/src/vlm/enrichment.py b/src/vlm/enrichment.py
index faefd19..b1eebf5 100644
--- a/src/vlm/enrichment.py
+++ b/src/vlm/enrichment.py
@@ -12,7 +12,7 @@ from urllib.request import urlopen, Request
 
 from vlm.cache import EnrichmentCache
 from vlm.config import Config
-from vlm.providers import ProviderResult, TMDBProvider, DoubanProvider
+from vlm.providers import ProviderResult, TMDBProvider
 from vlm.parser import normalize_title
 
 RefreshMode = str
@@ -156,6 +156,7 @@ def _emit_progress(stats: dict[str, int | list[dict[str, str]]], progress_callba
 
 def _build_providers(config: Config, *, request_timeout: int, retries: int) -> list:
     providers = []
+    unsupported: list[str] = []
     for name in config.enrichment_providers:
         key = name.lower()
         if key == "tmdb":
@@ -167,16 +168,14 @@ def _build_providers(config: Config, *, request_timeout: int, retries: int) -> l
                     min_interval_seconds=0.25,
                 )
             )
-        elif key == "douban":
-            providers.append(
-                DoubanProvider(
-                    config.douban_api_key,
-                    endpoint=config.douban_api_endpoint,
-                    timeout_seconds=request_timeout,
-                    retries=retries,
-                    min_interval_seconds=0.4,
-                )
-            )
+        else:
+            unsupported.append(name)
+
+    if unsupported:
+        raise ValueError(
+            f"Unsupported enrichment providers: {unsupported}. Supported providers: ['tmdb']"
+        )
+
     return providers
 
 
@@ -390,7 +389,7 @@ def _apply_payload(record: dict, payload: dict) -> None:
         "provider_metadata",
         "display_title",
     ):
-        if key in payload and payload[key] is not None:
+        if key in payload:
             record[key] = payload[key]
 
     if "needs_review" in payload:
@@ -422,10 +421,16 @@ def _fallback_title_from_filename(filename: Optional[str]) -> Optional[str]:
 
 
 def _build_display_title(record: dict, payload: dict, config: Config) -> str:
-    title_zh = payload.get("title_zh") or record.get("title")
-    title_en = payload.get("title_en") or record.get("title")
+    fallback_title = record.get("title")
+    title_zh = payload.get("title_zh") or fallback_title
+    title_en = payload.get("title_en") or fallback_title
+    if title_zh and title_en and title_zh == title_en:
+        title_en = ""
     try:
-        formatted = config.naming_title_format.format(title_zh=title_zh, title_en=title_en).strip()
+        formatted = config.naming_title_format.format(
+            title_zh=title_zh or "",
+            title_en=title_en or "",
+        ).strip()
     except Exception:
-        formatted = f"{title_zh} {title_en}".strip()
+        formatted = f"{title_zh or ''} {title_en or ''}".strip()
     return " ".join(formatted.split())
diff --git a/src/vlm/providers/__init__.py b/src/vlm/providers/__init__.py
index 395a8f7..c40cf0f 100644
--- a/src/vlm/providers/__init__.py
+++ b/src/vlm/providers/__init__.py
@@ -2,11 +2,9 @@
 
 from vlm.providers.base import EnrichmentProvider, ProviderResult
 from vlm.providers.tmdb import TMDBProvider
-from vlm.providers.douban import DoubanProvider
 
 __all__ = [
     "EnrichmentProvider",
     "ProviderResult",
     "TMDBProvider",
-    "DoubanProvider",
 ]
diff --git a/src/vlm/providers/douban.py b/src/vlm/providers/douban.py
deleted file mode 100644
index 104fa21..0000000
--- a/src/vlm/providers/douban.py
+++ /dev/null
@@ -1,97 +0,0 @@
-"""Douban provider implementation.
-
-This provider is optional. If no endpoint is configured, it silently degrades.
-"""
-
-from __future__ import annotations
-
-import json
-import time
-from typing import Optional
-from urllib.parse import urlencode
-from urllib.request import urlopen, Request
-
-from vlm.providers.base import ProviderResult
-
-
-class DoubanProvider:
-    """Fetch reputation data from a configurable Douban-compatible API."""
-
-    name = "douban"
-
-    def __init__(
-        self,
-        api_key: Optional[str],
-        endpoint: Optional[str] = None,
-        timeout_seconds: int = 6,
-        retries: int = 2,
-        min_interval_seconds: float = 0.4,
-    ) -> None:
-        self.api_key = api_key
-        self.endpoint = endpoint
-        self.timeout_seconds = timeout_seconds
-        self.retries = retries
-        self.min_interval_seconds = min_interval_seconds
-        self._last_request_at = 0.0
-
-    def enrich(self, *, title: str, media_type: str, year: Optional[int] = None) -> Optional[ProviderResult]:
-        if not self.endpoint:
-            return None
-
-        params = {
-            "q": title,
-            "type": media_type,
-        }
-        if year is not None:
-            params["year"] = year
-        if self.api_key:
-            params["api_key"] = self.api_key
-
-        data = self._get_json(self.endpoint, params)
-        if not data:
-            return None
-
-        items = data.get("items") or data.get("subjects") or []
-        if not items:
-            return None
-
-        item = items[0]
-        score = item.get("rating") or item.get("score")
-        votes = item.get("vote_count") or item.get("ratings_count")
-        title_zh = item.get("title")
-        title_en = item.get("original_title")
-
-        return ProviderResult(
-            provider=self.name,
-            canonical_id=f"douban:{item.get('id', 'unknown')}",
-            title_zh=title_zh,
-            title_en=title_en,
-            translation_source=self.name if title_zh or title_en else None,
-            reputation_score=float(score) if score is not None else None,
-            reputation_votes=int(votes) if votes is not None else None,
-            reputation_source=self.name,
-            raw_metadata={"id": str(item.get("id", ""))},
-        )
-
-    def _wait_for_rate_limit(self) -> None:
-        if self.min_interval_seconds <= 0:
-            return
-        now = time.monotonic()
-        elapsed = now - self._last_request_at
-        if elapsed < self.min_interval_seconds:
-            time.sleep(self.min_interval_seconds - elapsed)
-
-    def _get_json(self, url: str, params: dict) -> Optional[dict]:
-        full_url = f"{url}?{urlencode(params)}"
-        request = Request(full_url, headers={"Accept": "application/json"})
-        for _ in range(max(self.retries + 1, 1)):
-            self._wait_for_rate_limit()
-            try:
-                with urlopen(request, timeout=self.timeout_seconds) as response:
-                    payload = response.read().decode("utf-8")
-                    self._last_request_at = time.monotonic()
-                    return json.loads(payload)
-            except Exception:
-                self._last_request_at = time.monotonic()
-                continue
-        return None
diff --git a/tests/test_config.py b/tests/test_config.py
index 054f0b1..703f4c6 100644
--- a/tests/test_config.py
+++ b/tests/test_config.py
@@ -20,6 +20,7 @@ class TestConfig:
         assert config.series_template == "series/{title}/Season {season:02d}/"
         assert config.log_level == "INFO"
         assert config.quarantine_dir == ".quarantine"
+        assert config.enrichment_providers == ["tmdb"]
     
     def test_config_creation_with_custom_values(self):
         """Test creating Config with custom values."""
@@ -443,6 +444,15 @@ class TestValidateConfig:
         errors = validate_config(config)
         assert any("cannot be empty" in e for e in errors)
 
+    def test_validate_rejects_unsupported_enrichment_provider(self):
+        """Test validating config with unsupported enrichment provider."""
+        config = Config(
+            library_root=Path("/test"),
+            enrichment_providers=["tmdb", "douban"],
+        )
+        errors = validate_config(config)
+        assert any("unsupported providers" in e for e in errors)
+
     def test_validate_category_list_with_non_string(self):
         """Test validating config with non-string in category list."""
         config = Config(
diff --git a/tests/test_enrichment.py b/tests/test_enrichment.py
index 160fdb1..1725dc7 100644
--- a/tests/test_enrichment.py
+++ b/tests/test_enrichment.py
@@ -1,7 +1,9 @@
 """Unit tests for enrichment pipeline."""
 
+import pytest
+
 from vlm.config import Config
-from vlm.enrichment import enrich_identities_data
+from vlm.enrichment import _build_display_title, _build_providers, enrich_identities_data
 from vlm.providers.base import ProviderResult
 
 
@@ -169,3 +171,96 @@ def test_enrich_refresh_all_bypasses_cache(tmp_path, monkeypatch):
     _, stats = enrich_identities_data(identities, config, refresh_mode="refresh_all")
     assert provider.calls == 2
     assert stats["cache_hits"] == 0
+
+
+def test_build_providers_rejects_unknown_provider(tmp_path):
+    """Unknown providers should fail fast with a clear error."""
+    config = Config(
+        library_root=tmp_path,
+        enrichment_providers=["tmdb", "tmdb_typo"],
+    )
+
+    with pytest.raises(ValueError, match="Unsupported enrichment providers"):
+        _build_providers(config, request_timeout=3, retries=1)
+
+
+def test_refresh_all_clears_stale_enrichment_fields(tmp_path, monkeypatch):
+    """refresh_all with no new match should clear stale enrichment data."""
+
+    class FlakyProvider:
+        name = "dummy"
+
+        def __init__(self):
+            self.calls = 0
+
+        def enrich(self, *, title: str, media_type: str, year=None):
+            self.calls += 1
+            if self.calls == 1:
+                return ProviderResult(
+                    provider="dummy",
+                    canonical_id="dummy:test",
+                    title_zh="测试",
+                    title_en="Test",
+                    translation_source="dummy",
+                    reputation_score=8.8,
+                    reputation_votes=120,
+                    reputation_source="dummy",
+                )
+            return None
+
+    config = Config(
+        library_root=tmp_path,
+        enrichment_cache_db=tmp_path / "cache.db",
+        enrichment_providers=["dummy"],
+        translation_fallback_machine=False,
+    )
+
+    provider = FlakyProvider()
+    monkeypatch.setattr(
+        "vlm.enrichment._build_providers",
+        lambda _config, request_timeout, retries: [provider],
+    )
+
+    identities = {
+        "metadata": {},
+        "movies": [
+            {
+                "path": "/library/movie/Test.2020.mkv",
+                "filename": "Test.2020.mkv",
+                "category": "movie",
+                "title": "Test",
+                "year": 2020,
+                "confidence": 0.9,
+                "needs_review": False,
+            }
+        ],
+        "series": [],
+        "anime": [],
+        "other": [],
+    }
+
+    enrich_identities_data(identities, config, refresh_mode="incremental")
+    movie = identities["movies"][0]
+    assert movie["canonical_id"] == "dummy:test"
+    assert movie["title_zh"] == "测试"
+    assert movie["title_en"] == "Test"
+
+    enrich_identities_data(identities, config, refresh_mode="refresh_all")
+
+    assert movie["canonical_id"] is None
+    assert movie["title_zh"] is None
+    assert movie["title_en"] is None
+    assert movie["translation_source"] is None
+    assert movie["reputation_score"] is None
+    assert movie["reputation_votes"] is None
+    assert movie["reputation_source"] is None
+    assert movie["enrichment_confidence"] == 0.0
+
+
+def test_build_display_title_deduplicates_fallback_title(tmp_path):
+    """Fallback display title should not duplicate identical names."""
+    config = Config(library_root=tmp_path)
+    record = {"title": "Interstellar"}
+    payload = {"title_zh": None, "title_en": None}
+
+    assert _build_display_title(record, payload, config) == "Interstellar"